Data

Note: the max columns are NOT unique

Note: results columns are of type object since there's a timeout value. Remove these rows

Plot domain i.e. tunable search space

3d plot with two tunables on x and y axes and objective on z-axis

Note that all the highest and lowest objectives seem to be along the axes i.e. when one tunable is very low. Plot objective vs log difference

log difference was chosen because we want a metric that is sensitive whenever a tunable hits low values. The difference ensures that the limits: tunable 1 $\rightarrow$ 0 and tunable 2 $\rightarrow$ 0 get mapped to +/- $\infty$. This is not very carefully thought out and is just a simple transformation from a 2d space to a 1d space.

BayesOpt

Specifying the objective

Specifying the search space

run bayesopt

Search space